🟢 Fundamentals & First Steps

What it is

Almost every company wants to adopt AI, but a serious strategy consulting engagement costs thousands of reais, takes weeks, and depends on scarce experts. Most companies end up without a clear plan.

Why learn

Understanding the pain point shows the tool’s real value: it democratizes something that used to be exclusive to big budgets. You can now offer (or get) a diagnosis few people have.

Key concepts

• Traditional consulting = weeks + thousands of reais
• The tool = minutes + pennies
• Focus on people who normally wouldn’t have access

What it is

You provide a company name (and, if you like, some context), and the tool returns a complete AI strategy package—researched, written, and formatted automatically.

Why learn

This is the course's central mental model: a funnel from one simple input (the name) to many ready-to-use outputs (the documents). Everything else is a detail of that funnel.

Key concepts

• Minimum input = company name
• Optional context improves the result
• Output = ready-to-use package

What it is

There are 15 text documents (diagnostics, roadmap, ROI, governance, prompts, and more), 2 PowerPoint presentations, 2 Word reports, and architecture diagrams as images (PNG).

Why learn

Knowing what to expect prevents frustration and helps you plan how to use each piece. Track 3 breaks down each document—here, you just need the big picture.

Key concepts

• 15 documents + 2 PPTX + 2 DOCX + diagrams
• Everything saved in a folder for each company
• Presentation-ready material

What it is

The process has 3 sequential phases: Research (collects company data), Synthesis (writes the documents), and Generation (assembles presentations and diagrams). Each phase uses a different technology.

Why learn

Understanding the phases helps you track progress on the screen and know where to troubleshoot when something goes wrong (covered in Tracks 2 and 3).

Key concepts

• Phase 1 Research → uses Perplexity
• Phase 2 Synthesis → uses Gemini
• Phase 3 Generation → runs on your computer, at no cost

What it is

The tool combines two AI services: the Perplexity (searches the internet for up-to-date information, with sources) and the Google Gemini (turns the research into well-written documents).

Why learn

You'll need a key for each service. Knowing each AI's role clarifies why both are needed—and what each one charges.

Key concepts

• Perplexity = research with sources
• Gemini = document writing
• Each one requires its own API key

What it is

The only cost is for AI calls: about US$ 0,05 in quick mode and US$ 0,50 in in-depth mode, per company. The generation phase (on your computer) is free.

Why learn

Takes the fear out of the cost — it’s incredibly cheap. But it’s good to understand that each analysis uses a little of the credit on your API keys.

Key concepts

• Quick: ~US$ 0,05 · in-depth: ~US$ 0,50
• Local generation = free
• You pay the APIs directly, not the course

What it is

Suitable for consultants who want to deliver faster, founders evaluating AI in their own business, innovation teams, and anyone studying AI strategy. No programming knowledge required.

Why learn

Helps you frame the use case: for clients, for the company itself, or for learning. The result is always a strong draft that a human reviews and personalizes.

Key concepts

• Consulting, internal use, or education
• No programming prerequisites
• Generates drafts for a human to refine

What it is

Three things: the Python 3.9 or higher installed, the two API keys (Perplexity and Gemini) and a terminal (Command Prompt on Windows; Terminal on Mac/Linux).

Why learn

Having the prerequisites ready before you start prevents getting stuck in the middle of installation — the most common mistake for beginners.

Key concepts

• Python 3.9+ (download from python.org/downloads)
• On Windows, select “Add Python to PATH” during installation
• Terminal open in the project folder

What it is

Go to perplexity.ai/settings/api, create an account, and click "Generate API Key". The key starts with pplx-.

Why learn

It’s one of the two required keys. Without it, the Research phase won’t run, and no documents will be generated.

Key concepts

• Site: perplexity.ai/settings/api
• Key format: pplx-...
• Store it in a safe place — it’s like a password

What it is

Go to aistudio.google.com/apikey, sign in with your Google account and click "Create API Key". The key starts with AIzaSy.

Why learn

This is the second required key. Without it, the Synthesis phase can't write the documents.

Key concepts

• Site: aistudio.google.com/apikey
• Key format: AIzaSy...
• Sign in with a regular Google account

What it is

After downloading the project, run python setup.py. The script creates the environment, installs the dependencies, and prepares the .env file for you.

Why learn

This is the simplest option if you don't want to deal with manual commands. One command and the environment is ready.

Key concepts

• 1 command does it all
• Creates the virtual environment and installs dependencies
• Generates a blank .env for you to fill in

What it is

If you prefer to control each step: create the environment with python3 -m venv venv, enable it, install it with pip install -r requirements.txt and copy the example file with cp .env.example .env.

Why learn

Useful when the automated process fails, or to understand what’s happening under the hood. “venv” is an isolated environment just for this project.

Key concepts

• Enable: source venv/bin/activate (Mac/Linux)
• Or .\venv\Scripts\activate (Windows)
• Always activate the venv before running the tool

What it is

Open the file .env in a text editor and fill in PERPLEXITY_API_KEY e GEMINI_API_KEY with your keys. Save the file.

Why learn

This is where the keys are “fed” into the program. Without this, the tool can't access the AIs — and authentication fails.

Key concepts

• PERPLEXITY_API_KEY=pplx-...
• GEMINI_API_KEY=AIzaSy...
• No extra spaces · never share this file

What it is

Before spending anything, run a simulation test: python -m strategy_factory.main run "Stripe" --dry-run. If it runs without a key error, everything is fine.

Why learn

Confirms the installation and key configuration without spending a cent, because simulation mode doesn’t make real AI calls.

Key concepts

• --dry-run = simulation, no AI calls
• A key error here indicates an incorrect .env
• Remember to activate the venv first

What it is

With the environment activated, run python -m strategy_factory.webapp. The browser opens automatically at http://localhost:8888.

Why learn

The web mode is the most user-friendly way to use the tool — without having to memorize any command beyond this one.

Key concepts

• Command: ...strategy_factory.webapp
• Opens at localhost:8888
• Port in use? It finds another one, or use --port 9000

What it is

The screen has a field for the company name, a mode selector (Quick/Comprehensive), an optional context field, and a button to start the analysis.

Why learn

Getting to know the screen before you run it helps you feel confident about your first analysis—without worrying about clicking the wrong place.

Key concepts

• Company field (required)
• Mode: quick or in-depth
• Context (optional) + start button

What it is

Enter the name of the company you want to analyze (for example, “Stripe”). It’s the only information required to get started.

Why learn

This is the starting point for every analysis. The name also becomes the name of the folder where the results will be saved.

Key concepts

• Name = main input
• Turn into the results folder (e.g., stripe)
• Use the company’s commonly known name

What it is

In the context field, you can add hints like "fintech, B2B, 200 employees." This helps Perplexity research more precisely.

Why learn

Good context goes a long way toward improving document quality, especially for companies that aren’t well known or have ambiguous names.

Key concepts

• Context = industry, company size, business model
• Optional, but recommended
• The more specific you are, the better the result

What it is

After starting, a bar shows real-time progress through the 3 phases (Research, Synthesis, Generation), indicating what's happening.

Why learn

You see what's happening and how much time is left—and it's clear when the package is ready.

Key concepts

• Progress bar by phase
• Quick: 2-3 min · in-depth: 5-10 min
• Keep track without closing the tab

What it is

When you’re done, the screen shows links to download each deliverable (documents, presentations, Word files, and diagrams). Everything is also saved in the folder output/ from your machine.

Why learn

This is the end goal: having the files on hand to use, edit, and present. Track 3 shows you what to do with each one.

Key concepts

• Download each file directly from the screen
• Everything saved in output/nome-da-empresa/
• You can read the markdown in the interface itself